← Back to list
Hosking PartnersReport7 Nov 2023Source: hoskingpartners.com

Dark Side of the Sun

Hosking Partners is a London boutique founded in 2013 by Jeremy Hosking, a portfolio manager at Marathon Asset Management for over 25 years. It runs a single global equity strategy built on the capital-cycle, supply-side approach — contrarian, long-term, and unusually diversified (350+ holdings) under a multi-counsellor model, managing around $5.5bn.

Jeremy Hosking · 2013 · 伦敦Capital cycle / contrarian

In plain words

This report warns that China's solar industry faces deep problems: its cost advantage relies on cheap coal power and alleged forced labor, both unsustainable. Overcapacity and grid bottlenecks also threaten returns. The fund sold its stake in LONGi Green Energy in August 2023 because it couldn't verify forced labor risks in its supply chain. Instead, it favors Ferroglobe, a silicon metal producer that benefits from US policies (IRA) challenging China's dominance. Ferroglobe focuses on paying down debt and returning cash to shareholders, with potential to enter the solar market.

AI SummaryAI-generated · may contain errors · verify against the original

At a Glance

One-sentence summary of the author's market view this period: China's solar industry faces deep tensions between environmental (E) and social (S) issues. The overall industry outlook is cautious, but structural opportunities exist in non-China supply chains. [Cautious]

  • China's solar industry suffers from severe overcapacity (capacity in 2022 was more than double demand), compounded by global grid bottlenecks (approximately 3,000 GW of renewable energy capacity idle), leading to a bleak outlook for investment returns.
  • The cheap coal-fired power subsidies underpinning China's cost advantage are unsustainable, and future costs face reflationary pressures (a 1% rise in discount rates increases solar power costs by 6%).
  • Hosking Partners liquidated its position in Longi Green Energy in August 2023 due to the inability to confirm forced labor risks in the supply chain.
  • The U.S. Inflation Reduction Act (IRA) is reshaping the global competitive landscape. As the West challenges China's dominance, companies such as Ferroglobe hold structural advantages.
~19 min full read · 17 sections
Deep Analysis

China Solar Industry: Tensions Between Environmental and Social Issues

The report argues that China's solar industry is a classic case study of the complex relationship between environmental (E) and social (S) issues. The author believes that understanding this conflict is crucial for long-term investors, both to adjust portfolios to outperform during the energy transition and to guide engagement with portfolio companies. Over the past two years, the institution has explored several related themes through its Active Ownership Report, including the impact of developed-world decarbonization on the emerging world, the relationship between net-zero pathways and the social role of energy, and the influence of the energy transition on Russian geopolitics. These themes span industries and regions, while this report focuses on a more specific example—China's solar industry.

The author states: "The solar industry is, at its heart, a mining industry." This means: "The solar industry is essentially a mining industry." The report details the complete value chain from quartz sand to final solar panels: approximately 10 kilograms of quartz sand can produce 1 kilogram of photovoltaic silicon, which then goes through stages such as wafers, cells, and modules to become the final product. China currently dominates this value chain, producing 90% of the world's polysilicon, 96% of wafers, 83% of cells, and 75% of modules. The core driver of this dominance is government subsidies, particularly energy subsidies in the form of cheap coal-fired electricity.

Subsidy-Driven Cost Advantage and Price Distortion

The author argues that cheap coal-power subsidies are the biggest driver of the dramatic cost decline in China's solar industry, but current prices are already below the level needed for an economic return. The report points out that approximately 80% of the cash cost of photovoltaic silicon is energy cost. China's subsidized coal-power price is about 2–3 cents/kWh, roughly half the price of natural gas in the United States and about five times cheaper than long-term electricity prices in Europe. This cheap energy has enabled China to scale up production massively, monopolize the low end of the solar cost curve, and rapidly capture market share. In 2012, about half of solar project costs came from manufacturing; today, that share is only one-fifth.

The author states: "This manufacturing scale, facilitated by subsidised coal power, has been the largest driver of the remarkable deflation in solar costs witnessed in the past decade." This means: "This manufacturing scale, driven by subsidized coal power, has been the biggest driver of the remarkable deflation in solar costs seen over the past decade." China has driven photovoltaic silicon prices down to about $8–10/kg, but a bottom-up analysis shows that even with subsidies, a price of at least $12.50/kg is needed to generate a meager internal rate of return (IRR) of 5%. One study estimates that without any subsidies, China's photovoltaic silicon costs could be as high as $70/kg. This suggests that current prices may already be below economically sustainable levels.

Xinjiang Labor Controversy: A More Sensitive Subsidy Source

The report notes that even more controversial than coal-power subsidies is the alleged use of forced labor from the Uyghur ethnic group in Xinjiang within the solar value chain. The author describes this allegation as a "more sensitive source of subsidy," implying it may constitute another form of cost advantage. Although this chapter does not elaborate on details, combined with the full report overview, more than half of polysilicon is produced in Xinjiang, involving the so-called "transfer program" (up to 2.6 million people) allegedly linked to forced labor. This creates a core tension between the environment (cheap coal power driving solar adoption) and society (labor rights).

Investment Implications

The report clearly positions China's solar industry as a classic case of E-S tension, but this chapter does not provide specific investment action recommendations. Investors should note that, as a position-holding institution, its arguments may carry a bias toward justifying the holding. The key risk is that if labor controversies trigger trade restrictions or sanctions, the cost advantage of Chinese solar manufacturers could be eroded, thereby affecting the pricing and structure of the global solar supply chain.


Xinjiang Polysilicon Forced Labor Allegations Are a Known Unknown Risk

The report notes that 90% of the world's solar-grade polysilicon is manufactured in China, with over half originating in Xinjiang province, involving a "labor transfer" program that allegedly forces Uyghur Muslims to work. The author states: "Of the 90% of the world’s solar-grade polysilicon that is manufactured in China, over half originates in Xinjiang province." This region is the primary home of the Muslim Uyghur population. The Chinese government claims these transfers are voluntary, but evidence suggests they are compulsory, with up to 2.6 million people unable to refuse or leave. The United Nations International Labour Organization defines this as forced labor. The cotton and tomato harvesting industries are the largest recipients, but the solar industry also faces serious allegations, with upstream quartz mines and silicon smelting being most implicated due to their reliance on manual and low-skilled labor. However, due to the highly integrated nature of the industry, it is difficult to determine the extent of each company's involvement in the supply chain.

Western Analysts Were Bullish on Chinese Solar in 2020-21, but Hosking Has Exited LONGi

Despite these concerns, many Western analysts in 2020-21 viewed Chinese solar as a value and opportunity area, and Hosking Partners established a small position in LONGi Green Energy in July 2021, ultimately liquidating it in August 2023. The author argues that Chinese solar at the time boasted strong profit margins, market share, cost leadership, and a technological R&D moat. Compared to wind power, solar has the potential for continuous efficiency improvements and cost reductions without encountering engineering problems constrained by hard physical laws. The analogy to Taiwan's booming semiconductor industry in the early 2000s seemed reasonable. LONGi, focused on solar module manufacturing, is a technology leader with no production bases in Xinjiang and claims to require written commitments against forced labor from its 150 suppliers. In September 2021, Hosking began engaging with LONGi, encouraging greater transparency.

Engagement Failed to Clarify Forced Labor Risk, Combined with Supply-Side Deterioration Leading to Exit

Hosking's engagement efforts—including background research, management calls, and formal letters—failed to determine LONGi's exposure to forced labor risks, ultimately leading to the sale of the position in August 2023. The author notes that they inquired about how LONGi manages forced labor risks and the impact of U.S. sanctions on revenue, encouraging greater supply chain transparency (citing, for example, Associated British Foods' Primark supply chain mapping and human rights audit data). However, progress was minimal: LONGi appeared unable to fully engage on the issue because the Chinese government effectively prohibits companies from acknowledging the existence of forced labor among Uyghurs. LONGi was also reluctant to challenge its key suppliers—particularly Daqo New Energy and Hoshine Silicon Industry—which form long-term strategic partnerships through a complex network of joint ventures. The author concludes: "Our inability to encourage greater transparency on the materiality of the forced labour risk – combined with a deteriorating and related supply-side picture described in more detail below – led us to sell the portfolio’s position in LONGi in August 2023."

Investment Implications

This chapter reveals the tension between environmental (E) and social (S) issues in China's solar supply chain: low-cost advantages (reliant on coal-fired power subsidies) coexist with forced labor allegations. As a position holder, the author's decision to exit reflects the irreconcilability of governance (G) risks (government suppression of transparency) and supply chain risks (supplier linkages). Readers should note that this is a decision made by an active fund at a specific point in time based on limited information, not a judgment on the industry as a whole.


Here is the translation of the Chinese investment research notes into natural, professional English.


This is an analysis of the "Part 3" chapter of the Hosking Partners report.

At a Glance

Overcapacity and Grid Bottlenecks Are the Two Core Risks for the Solar Industry

The report argues that the Chinese solar industry faces a bleak outlook for investment returns, caught between severe overcapacity and global grid bottlenecks. The author notes that despite optimistic forecasts from sell-side analysts, the performance of Chinese solar stocks has been disappointing since mid-2021. The core reason is that the continuous expansion of upstream polysilicon capacity has led to a permanent state of oversupply.

  • Capacity Far Exceeds Demand: Data from the International Energy Agency (IEA) shows that manufacturing capacity, from polysilicon to modules, exceeds demand, and the surplus is worsening. In 2022, capacity was more than double the demand (see Figure 2 in the original report).
  • Declining Capacity Utilization: Due to improved polysilicon efficiency per wafer, polysilicon capacity utilization has fallen from 85% in 2004 to the current 60%.
  • Expansion Continues: Although financing conditions have tightened (e.g., producer Tongwei Co., Ltd. was recently forced to cancel an equity financing plan), the top eight Chinese producers still have sufficient cash to support their full expansion plans through 2026.
  • Grid Bottlenecks: Globally, approximately 3,000 GW of renewable energy capacity is idle due to insufficient grid construction, equivalent to five times the total global new solar installations in 2022. The author uses an adapted proverb to describe this situation: "'to put the solar panel before the grid!'"

Cost Advantages Are Unsustainable, Facing Future Reflationary Pressures

The report warns that the subsidies (cheap coal-fired power, labor, etc.) underpinning the cost advantages of the Chinese solar industry are unsustainable, and costs may face upward pressure in the future. The author believes that cost reductions achieved through economies of scale are nearing their end, and future cost declines will depend more on materials, cell efficiency, and capital costs, all of which are subject to uncertainty.

  • Unsustainability of Subsidies: The author states: "These are ultimately unsustainable (in several senses of the word), and as such the profits of each industry player need to be discounted heavily in any valuation analysis."
  • Factors Driving Cost Increases: Research shows that for every 1% increase in the discount rate, solar power generation costs rise by 6%. At the same time, material bottlenecks across the value chain are tightening.
  • Geopolitical Impact: The U.S. import ban on modules manufactured in Xinjiang has resulted in modules equivalent to a full year of European demand (50 GW) being stranded in warehouses near Rotterdam. These inventories will eventually hit the market, causing a double blow of asset inflation and lower future profits.

Industry Consolidation Is a Distant Prospect, and the U.S. IRA Act Introduces New Variables

The report judges that the consolidation cycle for the Chinese solar industry is far from imminent, while the U.S. Inflation Reduction Act (IRA) is reshaping the global competitive landscape. Although analysts have predicted industry consolidation since 2012, the number of large Chinese manufacturers has increased rather than decreased.

  • Oligopoly Without Discipline: The world's largest solar module manufacturer alliance, the "Silicon Module Super League" (SMSL), has grown from 5 founding members in 2015 to 7 today, 6 of which are Chinese companies. The polysilicon sector is similar, with market share relatively evenly distributed among 6 large players. However, this oligopolistic structure has not brought capital discipline, and corporate capital expenditure often exceeds operating cash flow (see Figure 4 in the original report).
  • Impact of the U.S. IRA: Through the IRA, the U.S. has committed to a fiscal stimulus investment cycle for its domestic solar industry similar to China's in 2011. Additionally, starting in August 2024, the U.S. will expand tariffs on Chinese modules to products assembled in Southeast Asia.
  • Valuation Changes: These policies have led to a revaluation of Chinese solar stocks over the past year. The average price-to-earnings (P/E) ratio of the top five producers has halved from 31x in 2022 to a projected 16x for 2023. The author believes this is "cheaper, but still not cheap." Meanwhile, although subsidized coal-fired power is a significant cost advantage for Chinese producers, U.S. natural gas is a valuable substitute and is less susceptible to reflationary pressures from rising carbon prices (or carbon taxes) in Western markets.

Investment Implications

The report clearly expresses a cautious stance on the Chinese solar industry as a whole. Investors should note that this is an analysis from the perspective of an actively managed fund, with a clear value-investing and risk-aversion bias. The core implication of the report is: Under multiple pressures such as overcapacity, grid bottlenecks, fading cost advantages, and geopolitical risks, the profitability of Chinese solar manufacturers (such as Tongwei Co., Ltd., LONGi Green Energy, etc.) may face long-term headwinds, challenging their high-growth narrative. When evaluating this industry, investors need to give sufficient weight to these structural risks rather than focusing solely on the demand-side growth story.


3. Clear Investment Opportunities Exist in the Non-China Solar Supply Chain

Hosking Partners believes that in the process of the West challenging China's solar dominance (e.g., through the IRA), companies like Ferroglobe possess structural advantages. The author notes that the institution is skeptical about investing in the Chinese solar market but remains focused on opportunities arising from solar adoption, placing greater emphasis on measurable supply bottlenecks rather than speculative demand forecasts. The author's original statement, "it is measurable supply bottlenecks rather than conjectured demand forecasts which most pique our interest," translates to: "What most piques our interest is measurable supply bottlenecks, not conjectured demand forecasts."

  • Ferroglobe is the position in this portfolio that embodies this logic. It is one of the world's largest producers of silicon metal and related alloys, primarily supplying sub-solar-grade silicon to the chemical and aluminum industries, and ferrosilicon and manganese to the steel industry. The author emphasizes its vertical integration (from coal and quartz to silicon metal production) and global production scale, which provide advantages in costs, raw material prices, and tariffs, and it is typically a leader in all markets outside of China.
  • As the West challenges China's dominance in the solar value chain through the IRA, companies like Ferroglobe can supply silicon metal to Western upgrade manufacturers, leveraging existing assets to enter the non-China photovoltaic silicon market.
  • The author acknowledges the risks: persistent Chinese overcapacity is a threat to Ferroglobe's pricing power, and currently, only early signs of a potential slowdown in Chinese capacity growth are visible. However, in the long term, China's economic pain and excessive debt may pressure subsidies, which could actually benefit Western competitors—who have already undergone consolidation during the past decade of China's cost leadership cycle—while Ferroglobe's healthy balance sheet lays the foundation for future returns.

4. Ferroglobe's Capital Allocation and Growth Prospects: Focus on Deleveraging, Shareholder Returns, and Solar Opportunities

The author points out that Ferroglobe is not blindly investing in new assets but is instead focusing on deleveraging and shareholder returns, while exploring solar opportunities through strategic partnerships. The author's original statement, "Instead of splashing the cash on new assets, Ferroglobe are focusing on deleveraging and delivering returns to shareholders," translates to: "Ferroglobe is not splashing cash on new assets but is focusing on deleveraging and delivering returns to shareholders."

  • In terms of capital expenditure, the average cost for US polysilicon manufacturers to increase unit capacity remains more than four times that of China. Ferroglobe has chosen to establish strategic partnerships with specialized producers like REC Silicon to explore solar opportunities.
  • Looking ahead, by 2030, incremental solar-related silicon metal consumption in North America could reach one-third of current consumption, and this sub-sector alone could generate $300 million in EBITDA for Ferroglobe. For context, the adjusted EBITDA for the entire business in 2022 was $795 million (Source 13: Bloomberg, as of October 2, 2023).
  • The company is also involved in R&D technology for silicon cathodes in electric vehicle batteries, which could become an exciting source of future profits.
  • Additionally, Ferroglobe is not subject to the regulatory and reputational discounts associated with forced labor issues in China, a problem that shows no signs of easing in the near to medium term.

Investment Implications

Hosking Partners demonstrates its investment process through the Ferroglobe case: a combination of supply-side analysis, ESG factor consideration, and active engagement. As of October 2023, nearly 30% of the portfolio was allocated to sectors related to the energy transition, with a concentration in areas that have seen underinvestment recently and are undervalued (e.g., mining, long-life traditional energy assets, tanker shipping). The author believes that the Chinese solar industry is the clearest example of the tension between "E" and "S" in the energy transition, and the institution has largely avoided this bubble built on cheap energy and cheap capital. Institutional perspective bias: The author presents Ferroglobe as a positive case, but readers should note that this is a position-holder's perspective, and Chinese overcapacity remains a core risk to its pricing power.


Position Moves

Ticker Direction Author's One-Sentence View Key Data
LONGi Green Energy Liquidated Sold holdings due to inability to confirm forced labor risk exposure and deteriorating supply conditions Liquidated in August 2023; initial engagement began in September 2021, encouraging greater transparency
Ferroglobe Hold for Observation Bullish on its vertical integration, global scale, and potential to enter the non-China solar silicon market Incremental demand for solar-related silicon metal in North America could reach one-third of current consumption, generating $300 million in EBITDA; 2022 adjusted EBITDA was $795 million
Tongwei Co. Not Disclosed Capacity expansion continues, but financing conditions are tightening (recently canceled equity financing plans) The top eight Chinese producers still have sufficient cash to support all expansion plans through 2026
Daqo New Energy Not Disclosed As LONGi's primary supplier, forms a long-term strategic partnership through complex joint ventures LONGi is reluctant to question its suppliers because the Chinese government effectively prohibits companies from acknowledging the existence of forced labor
Hoshine Silicon Industry Not Disclosed Same as above; one of LONGi's major suppliers Same as above
REC Silicon Not Disclosed Ferroglobe's strategic partner, jointly exploring solar opportunities Unit capacity costs for U.S. polysilicon manufacturers are on average more than four times those of Chinese producers